Year : 2008 | Volume
: 51 | Issue : 3 | Page : 379--381
Evaluation of various screening tests to detect asymptomatic bacteriuria in pregnant women
J Jayalakshmi1, VS Jayaram2,
1 Department of Microbiology, PSG Institute of Medical Sciences and Research, Coimbatore, Tamil Nadu, India
2 Department of Microbiology, Dr BR Ambedkar Medical College, Bangalore, Karnataka, India
Department of Microbiology, PSG Institute of Medical Sciences and Research, Peelamedu, Coimbatore - 641 004, Tamil Nadu
To evaluate the diagnostic efficacy of various screening tests in detecting asymptomatic bacteriuria among pregnant women. Clean catch midstream urine specimens were collected from 630 consecutive pregnant women and processed. Forty-four (7.4%) of the urine samples were culture positive, with Escherichia coli as the predominant organism isolated (57.4%). The results of the four screening tests, viz., Gram«SQ»s staining of uncentrifuged urine, pus cell count, nitrite test and leukocyte esterase (LE) test, were compared against urine culture. Gram«SQ»s stain of the uncentrifuged urine was found to be the single most useful test with a sensitivity and negative predictive value (NPV) of 85.1% and 98.8%, respectively. Pus cell count was the least sensitive. Neither the nitrite test nor the LE test alone was sensitive enough with 74.4% and 61.7%, respectively. However, when either or both tests positive were considered, it increased the sensitivity and NPV comparable with Gram«SQ»s staining results, with 82.9% and 98.8%, respectively. With the potential to be used as an office diagnostic procedure, the combined nitrite and LE dipstick test may provide an acceptable alternative.
|How to cite this article:|
Jayalakshmi J, Jayaram V S. Evaluation of various screening tests to detect asymptomatic bacteriuria in pregnant women.Indian J Pathol Microbiol 2008;51:379-381
|How to cite this URL:|
Jayalakshmi J, Jayaram V S. Evaluation of various screening tests to detect asymptomatic bacteriuria in pregnant women. Indian J Pathol Microbiol [serial online] 2008 [cited 2019 Nov 19 ];51:379-381
Available from: http://www.ijpmonline.org/text.asp?2008/51/3/379/42516
Infection of the urinary tract is a common problem in pregnancy due to the morphological and physiological changes that takes place in the genitourinary tract during pregnancy. It may be of two types, symptomatic or asymptomatic. Asymptomatic bacteriuria (ASB) may be defined as the "presence of actively multiplying bacteria within the urinary tract excluding the distal urethra", at a time when the patient has no urinary tract symptoms. 
Pregnant women with ASB are more likely to develop acute pyelonephritis in later pregnancy, postpartum urinary tract infection, hypertensive disease of pregnancy, anemia, chronic renal failure, prematurity, low birth weight babies and prenatal death if untreated. The incidence of these can be decreased by treating ASB during pregnancy promptly. 
In view of the high prevalence of 2-11%  ASB and the potential adverse sequelae, most centers perform routine analysis of midstream urine specimen during one or more antenatal visit. However, full bacteriological analysis is both time-consuming and expensive and vast majority of antenatal urine specimen will be negative to culture.
Screening methods may be useful because full bacteriological analysis could be reserved for those patients who are symptomatic or have a positive screening test results. Because ASB is detectable and treatable, screening of urine during pregnancy may be justified.
Materials and Methods
This cross-sectional study was undertaken after obtaining Institutional Ethical Committee's approval. About 20 mL of clean catch midstream urine samples were obtained from 630 pregnant women attending the obstetric outpatient of KC General Hospital, Bangalore and Dr BR Ambedkar Medical College Hospital, Bangalore.
Samples thus collected were immediately transported to the Diagnostic Microbiology Department and processed within 1 h. In case of delay, the samples were refrigerated at 4°C. The urine specimens were first processed by routine quantitative culture and were then tested by the screening methods under consideration.
A semi-quantitative calibrated loop technique was adopted for the primary isolation of the organism. A loopful of well-mixed uncentrifuged urine was streaked on to the surface of Blood agar and CLED agar. After incubating aerobically for 24 h at 37°C, colony forming unit (CFU) per milliliter of urine was described. The significant bacterial isolates were identified by standard procedures and subjected to antibiotic susceptibility by Kirby Bauer's disc diffusion method.
Diagnostic criteria for ASB were considered when at least two consecutive urine specimens showed ≥10 5 CFU/mL of urine of the same single species in the absence of symptoms.
Gram's staining of uncentrifuged urine
A drop of uncentrifuged well mixed urine was taken on a clean grease-free slide and stained by Gram's method of staining and examined under the oil immersion objective of the microscope (examining 20 fields). Presence of ≥1 bacteria per oil immersion field correlates with significant bacteriuria of ≥10 5 CFU/mL of urine.
Pus cell count of the uncentrifuged urine
To estimate pyuria, a Neubauer's counting chamber was used, a pus cell count of ≥10 cells per microliter of urine corresponds to an excrement rate of 4 × 10 5 leukocytes per hour, which was considered significant.
Leukocyte esterase test and Nitrite test
Uncentrifuged urine specimens were tested with colorimetric Combur-10 multireagent test (Boehringer Mannheim and Co.), for the presence of nitrite and leukocyte esterase activity, following the manufacturer's instructions.
Out of the total number of 630 pregnant women included in this prospective study, 47 (7.4%) were identified by culture to have significant bacteriuria. Escherichia coli was the predominant organism isolated (57.4%) followed by other uropathogens such as Klebsiella pneumoniae (19.21%), Staphylococcus saprophyticus (10.60%), Citrobacter diversus (4.25%), Enterococcus faecalis (2.12%), Proteus mirabilis (4.25%) and Enterobacter cloacae (2.12%).
Culture was taken as the gold standard against which the screening tests were compared. Using statistical software SSPS - 11 version, the statistical analyses were done and their sensitivity, specificity, positive predictive value (PPV) and negative predictive values (NPV) were derived. Maximum number of true positives were identified by Gram's staining of uncentrifuged urine (40/47), increasing the sensitivity and least number of positives identified by pus cell count (20/47), thereby decreasing the sensitivity [Table 1]. High false positives were seen with LE test and low false positives with nitrite tests, thereby decreasing and increasing the specificity of LE test and nitrite test, respectively, as shown in [Table 1]. The sensitivity (85.1%) and NPV (98.8%) of Gram's stains were found to be much superior to the others. However, nitrite test gave higher specificity (99.6%) and PPV (94.5%) compared to the rest [Table 2].
Since nitrite and leukocyte esterase combined test could be performed with a single dipstick test, statistical values were derived, considering both test results. The sensitivity and NPV increased to 82.9% and 98.8%, respectively when either nitrite or leukocyte esterase positive and both tests positive were considered, instead of single test values. Combined nitrite and leukocyte esterase test, however, gave low sensitivity (53.1%) because only 25 cases were positive for both tests. Since no false positives were wrongly identified by the combined nitrite and leukocyte esterase test, it gave a 100% PPV and specificity [Table 2].
Twenty-six out of the 630 urine samples were positive for pus cell counts with more than 10 pus cells per microliter of urine with culture being negative, this constituted 4.1% of sterile pyuria. Only 42.5% (20/47) ASB patients had pyuria associated with bacteriuria.
An early detection and treatment ASB may be of considerable importance not only to forestall acute pyelonephritis and chronic renal failure in the mother, but also to reduce prematurity and fetal mortality in the offspring.
An ideal screening test should be simple, rapid and accurate and must identify all positive cases, thus a sensitive test with a high NPV and specificity is desirable. Gram's stain of uncentrifuged urine was found to be the most useful single test with good sensitivity (85.1%), specificity (98.9%) and NPV (98.8%), when only those urine containing a pure culture of a potential pathogen of more than 10 5 CFU/mL of urine were considered as positive. It, however, requires trained personnel to screen and the results are subjected to observer variation. 
Although the pus cell count of unspun urine using a Neubauer's counting chamber is a very accurate method, it is very cumbersome and in our study gave low sensitivity and NPV, therefore not useful for screening ASB. The low sensitivity for pyuria observed suggests most probably but not always due to bladder colonization rather than actual infection. However, this is controversial.  Besides, hypotonic urine or alkaline urine due to the presence of Proteus , Klebsiella and pseudomonas can cause disintegration of the pus cells. The prevalence of 4.1% sterile pyuria may be attributed to infections due to organisms like Chlamydiae , which fail to grow in the media used for isolation. 
The nitrite test is an indirect measure of nitrate reducing bacteria, which includes all enterobacteriaceae, most non-fermenters and Gram-negative cocci, provided the urine contains sufficient dietary nitrates and has been retained in the bladder for longer than 4 h. The first voided urine specimen has been proved to be accurate, but collection was not practical in all our patient population and similar difficulties were addressed by previous workers. , The nitrite test gave lesser sensitivity (74.4%) as 12 positive cases were missed as false negatives that include all infections caused by Gram-positive cocci.
Although some investigators have reported dipstick LE sensitivity to be 100%,  other workers show much lower values. ,, The present study gave a lower sensitivity (61.7%) and NPV (96.7%) because of 42 false positives, which included 26 patients with sterile pyuria, identified, the specificity was lower (92.7%) than the other tests. False positive findings may result from high urinary levels of ascorbic acid, phenazopyridine and albumin. However, the LE test had the advantage to identify ASB even when the pus cells were disintegrated in the specimen.
Among the screening tests evaluated, Gram's staining of uncentrifuged urine was the best and pus cell count was found not useful. Neither the nitrite test nor the LE test was acceptable by itself as a screening test. However, when either nitrite test or LE test positive was considered, the sensitivity results were comparable to that of the Gram's stain results. The dipstick test with both the parameters had the advantage of being simple, rapid, not requiring any expertise or equipment and costing 1/10 th of culture.  Therefore, dipstick could be used as an office diagnostic test to screen ASB among pregnant women and reserve routine culture only for those screened positive.
|1||Hankins GD, Whalley PJ. Acute urinary tract infections in pregnancy. Clin Obstet Gynaecol 1985;28:266-78.|
|2||Delzell JE, Lefevre ML. Urinary tract infections during pregnancy. Am Fam Physician 2000;61:713-21.|
|3||Etherington J, James DK. Reagent strip testing of antenatal urine specimens for infection. Br J Obstet Gynaecol 1993;100:806-8.|
|4||Mathew J, George S, Mathews P, Mathai E, Brahmadattan KN, Seshadri L. The Griess test: An inexpensive screening test for asymptomatic bacteriuria in pregnancy. Aust NZ Obstet Gynecol 1998;69:550 -2.|
|5||Komaroof AL. Acute dysuria in women. N Engl J Med 1984;310:368-75.|
|6||Nuns D, Smith AR, Hosker G. Reagent strip testing of urine for significant bacteriuria in a urodynamic clinic. Br J Urol 1995;76:87-9.|
|7||Robertson AW, Duff P. The nitrite and leucocyte esterase for the evaluation of asymptomatic bacteriuria in obstetric patients. Obstet Gynecol 1988;71:878-81.|
|8||Hagay Z, Levy R, Miskin A, Milman D, Sharabi H, Insler V. Uriscreen, a rapid enzymatic urine screening test: Useful predictor of significant bacteriuria in pregnancy. Obstet Gynecol 1996;87:410-3.|